1. The Field of the Invention
Embodiments of the present invention relate to methods and apparatus for reducing back reflections into the source laser in a fiber-optic network.
2. Description of the Related Art
In the field of data transmission, one method of efficiently transporting data is through the use of fiber-optics. Digital data is propagated through a fiber-optic cable using light-emitting diodes or lasers. Light signals allow for extremely high transmission rates and very high bandwidth capabilities. Also, light signals are resistant to electromagnetic interferences that would otherwise interfere with electrical signals. Light signals are more secure because they do not allow portions of the signal to escape from the fiber-optic cable as can occur with electrical signals in wire-based systems. Light also can be conducted over greater distances without the signal loss typically associated with electrical signals on copper wire.
One goal in modern fiber-optic communication configurations is to maintain the integrity of the signal generated by the laser or the light emitting diode. One common problem that degrades the integrity of the signal generated occurs when portions of the signal are reflected back into the laser. Such reflections can cause increased transmission noise or bit error rates, increased laser noise due to reflections that cause optical resonance in the laser, and other similar problems.
One cause of this type of signal reflection occurs when the laser beam signal leaves a medium having a first index of refraction, and then enters a medium having a second and different index of refraction. For example, when a Distributed Feedback (DFB) laser is interfaced with a fiber-optic pigtail with free space between the laser and the receiving end of the pigtail, reflections often occur. Reflections of the laser beam back into the laser (or similar light source) are commonly referred to as “back reflections.”
Back reflections are commonly measured in terms of a ratio of the amount of the laser beam that is reflected compared to the transmitted part of the laser beam. This value is commonly expressed as a logarithmic ratio. In terms of this logarithmic ratio, DFB lasers commonly require back reflection levels as low as −40 dB to operate properly. One specific type of reflection that needs attention is “near-end back reflection.” A near-end back reflection is one caused by the first one or two connections from a laser transceiver to a fiber-optic pigtail and to a communications panel. Because these first connections generally occur in a fiber-optic cable that is not subjected to bending and heat stresses, the state of polarization of the laser beam can be predicted fairly accurately.
One example of a method for controlling near-end back reflections is shown in FIG. 1, which generally shows a Transmitter Optical Subassembly (TOSA) designated generally as 100. The TOSA 100 has a DFB laser 102 coupled to an optical isolator 104. The optical isolator 104 includes a 0° polarizer 106 coupled to a Faraday rotator 108, which is in turn coupled to a 45° polarizer 110. The DFB laser 102 emits a beam 114 which may be of any polarization as illustrated by the polarization indicator 112. The beam 114 passes through the 0° polarizer 106, which allows only the portions of the beam polarized at 0° to pass through, causing the beam 114 to be polarized at 0° as shown by the polarization indicator 116.
The beam 114 then passes through the Faraday rotator 108, which, in this example, is designed to rotate the beam 114 by 45° in the positive direction. The Faraday rotator 108 may be formed from a latching magnetic material or non-latching magnetic material. For a non-latching material, an external magnet 109 may be used to apply a magnetic field, while a latching material does not need an external magnetic field. This rotation causes the beam 114 to be polarized at 45° as is shown by the polarization indicator 118.
The beam 114 then passes through the 45° polarizer 110 without disruption as the optical axis of the 45° polarizer 110 and the polarization of the beam 114 are aligned. The beam 114 remains polarized at 45° as is shown by the polarization indicator 120. The beam 114 is then propagated through an air space 122 into a fiber-optic pigtail 124. While shown as a single discrete fiber-optic component, the fiber-optic pigtail actually represents all of the various connections that may exist in a fiber-optic network that cause back reflections including pigtail, transceiver, communication panel, and other connections.
Due to the difference in the index of refraction of the fiber-optic pigtail 124 (for example, about 1.47) and the air space 122 (about 1.0) at various connections within the network, a reflected beam 126 is often propagated back towards the DFB laser 102. The reflected beam 126 (or back reflection) may have any state of polarization (as shown by the polarization indicator 128), since the reflected beam 126 is caused by various components within the network. However, a major part of the reflected beam 126 is the near-end reflection caused by the first few components into which the beam 114 is transmitted. If these components are not subjected to mechanical and thermal stress, these near-end portions of the reflected beam are typically polarized at approximately 45°.
The reflected beam 126 passes through the 45° polarizer 110 such that only the portions of the reflected beam 126 that are polarized at 45° are allowed to pass through. This causes the reflected beam 126 to be polarized at 45° as shown by the polarization indicator 130. The reflected beam 126 then passes through the Faraday rotator 108, where it is rotated by positive 45° such that it is polarized to 90° as shown by the polarization indicator 132. Note that the Faraday rotator 108 rotates all beams passing through the Faraday rotator 108 by positive 45° irrespective of the direction of travel. The reflected beam 126 polarized at 90° has no 0° components and is therefore totally rejected from passing through the 0° polarizer 106. In this way, back reflections into the DFB laser 102 are minimized.
While in theory this method appears to completely block any back reflections into the DFB laser 102, in practice this often may not be the result. An ideal polarizer only allows beams to pass through at the angle of polarization of the polarizer. However, actual polarizers allow small portions of the beam perpendicular to the angle of polarization to leak through (referred to as leakage). One characteristic that determines the quality and often the price of a polarizer is the polarizer's ability to minimize the leakage of perpendicular polarization components of beams passing through the polarizer. This characteristic is known as the polarizer's “extinction ratio.”
Commonly, conventional polarizers used in the TOSA 100 environment described above have a perpendicular beam extinction ratio (or simply “extinction ratio”) of about −40 to −45 dB. One such polarizer is known as a pulled-glass polarizer. While using such polarizers effectively meets the operating criteria for most DFB lasers, such polarizers can be relatively expensive. In fact, the polarizers represent as much as 70% of the total isolator cost, and thus can significantly increase the overall relative cost of the isolator component. It would therefore be beneficial to construct an optical isolator using polarizers that are less expensive, thereby reducing the overall cost of the optical component.